Genomic and proteomic approaches to understand oral cancer Although risk factors for HNSCC, such as alcohol and tobacco consumption, are well recognized, the molecular mechanisms responsible for this malignancy are still not fully understood. We have used a number of novel approaches to investigate gene and protein expression profiles in HNSCC. We have shown that laser capture microdissection (LCM) can be used to procure specific cell populations from heterogenous tumor samples, and that LCM-procured material can be used effectively to extract RNA, DNA, and proteins. We have teamed up with other research institutions to conduct gene and protein expression analysis of HNSCC by combining LCM, gene arrays, and proteomic platforms. These efforts have already provided a wealth of information about the distinctive pattern of gene and protein expression in HNSCC. Gene and protein expression analysis: We have recently asked whether the differences in etiological factors associated with HNSCC, such as smoking and alcohol consumption in the US and other Western countries and betel quid chewing in South and Southeast Asia, result in the expression of a distinct set of genes that may contribute differentially to tumor progression. We found numerous genes that were differentially expressed between normal oral mucosa and cancer, and many distinct genes when comparing cancers associated with betel quid chewing and tobacco use. This finding further support the existence of distinct genetic and epigenetic changes during their malignant progression, which may impact the prognosis and treatment response. This body of information enabled us to identify a large number of molecules whose contributions to HNSCC progression are now under investigation in our laboratory and in the extramural community. Examples of molecules that were evaluated during this reporting period for their correlation with disease progression and/or contribution to HNSCC cell growth and invasive potential include sphingosine kinase, vimentin, EPS8, and FOXM1. Dysregulated signaling networks in HNSCC: novel mechanism-based approaches for HSNCC treatment There is an urgent need for new treatment options for HNSCC patients, considering that their overall 5-year survival is relatively low (50%) and has not improved much over the past 3 decades. The emerging information on the nature of the deregulated molecular mechanisms responsible for HNSCC progression has provided the possibility of exploring new mechanisms-based therapeutic approaches for HNSCC. For example, we have observed that persistent activation of the serine-threonine kinases mTOR is frequent event in HNSCC, and that inhibition of mTOR by the use of rapamycin causes the rapid decrease in the level of pS6 (a downstream target of mTOR) and the apoptotic death of HNSCC tumor xenografts, thereby causing tumor regression. These efforts have identified the Akt-mTOR pathway as a potential therapeutic target for HNSCC. Recent studies suggest that hypoxia, a common tumor microenvironmental stress, blocks the Akt-mTOR mitogenic pathway. Indeed, we have recently shown that hypoxia causes decreased levels of phosphorylation of the mTOR downstream target S6 (pS6) in HNSCC cells. This was associated with a marked upregulation of REDD1, a recently identified hypoxia-induced protein that negatively controls mTOR activity. Conversely, pS6 levels were retained under hypoxia in REDD1 knock-down cells and in HNSCC cells lacking endogenous REDD1 expression. Furthermore, we observed that prolonged hypoxia induced an energy-depleting response involving decreased cellular ATP levels and AMP-activated protein kinase (AMPK) activation. Interestingly, AMPK inhibition prior to prolonged hypoxia prevented REDD1 expression, thereby sustaining mTOR activity. These results suggest a novel mechanism by which AMPK activation following hypoxia-induced energy stress may be crucial in regulating REDD1 expression to control the mTOR pathway in HNSCC. Furthermore, we found that in many HNSCC cells the reduced mTOR activity in response to hypoxia via AMPK/REDD1 was deregulated, which hence might contribute to the persistent activation of the mTOR pathway in this cancer type. In this reporting period we have made a concerted effort to use both genetically-defined and conventional and novel chemical carcinogenesis models to evaluate the effectiveness of mTOR inhibitors for the treatment of HNSCC. For example, we have recently observed that rapamycin exerts a remarkable anti-cancer activity in a well established skin chemically-induced cancer model, decreasing the tumor burden of mice harboring early and advanced tumor lesions, and even recurrent skin SCCs. Immunohistochemical studies on tumor biopsies and clustering analysis revealed that rapamycin causes the rapid decrease in the phosphorylation status of mTOR targets, followed by the apoptotic death of cancer cells and the reduction in the growth and metabolic activity of the surviving ones, concomitant with a decrease in the population of cancer cells expressing mutant p53. Together with the potent antitumoral activity of rapamycin analogs in numerous solid tumors, and the recent approval by the FDA of their use in advanced renal carcinoma, our findings provided a strong rationale for the early evaluation of mTOR inhibitors as a molecular targeted approach to treat HNSCC. Animal models for oral malignancies A major limitation in the area of HNSCC research is the limited availability of animal models to test the validity of current genetic paradigms of tumorigenesis, and to explore the effectiveness of treatment modalities or chemopreventive approaches. A systematic analysis of available Cre recombinase systems for their ability to enable the inducible development of oral malignancies led us to establish a suitable and rapid oral-specific animal tumor model system. This system consists in the use of mice expressing a tamoxifen-inducible Cre recombinase under the control of the K14 promoter (K14-CreERTAM) and mice in which the ras oncogene is expressed from its own promoter after Cre excision of a transcription stop signal. These mice develop large tumors exclusively in the oral cavity within 1 month of a brief tamoxifen treatment, and if these mice are crossed with floxed-p53 conditional knock out mice, the compound mice develop carcinomas exclusively on the tongue as early as 2 weeks after tamoxifen induction. A remarkable achievement during this reporting period was the development of an oral-specific chemical carcinogenesis model recapitulating human HNSCC. We showed that the administration in the drinking water to mice of 4-Nitroquinoline-1 oxide (4NQO), a DNA adduct-forming agent that serves as a surrogate of tobacco exposure, leads to the progressive appearance of preneoplastic and tumoral lesions in the tongue and oral mucosa, with 100% incidence after only 16 weeks of carcinogen exposure. Remarkably, many of these lesions evolve spontaneously into highly malignant SCCs few weeks after 4NQO withdrawal. In this model, we have observed that the activation of the Akt-mTOR biochemical route represents an early event, already detectable in dysplastic lesions. Inhibition of mTOR by the chronic administration of rapamycin halts the malignant conversion of precancerous lesions and promotes the regression of advanced carcinogen-induced SCCs. This study provided a quite facile oral-specific chemical carcinogenesis model system for the evaluation of molecular targeted approaches for oral cancer treatment. This study may also enable the development of mouse models that incorporate both genetic engineering and 4NQO-induced carcinogenesis for future studies of specific molecular events that contribute to HNSCC initiation and progression.